Feature selection methods in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the SelectKBest feature selector from scikit-learn.

ML Python

from sklearn.feature_selection import [1]

Drag options to blanks, or click blank then click option'

AStandardScaler

BPCA

Ctrain_test_split

DSelectKBest

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing PCA which is for dimensionality reduction, not feature selection.

Using train_test_split which is for splitting data, not feature selection.

The SelectKBest class is used for feature selection by selecting the top k features based on a scoring function.

2fill in blank

medium

Complete the code to select the top 5 features using SelectKBest with the chi2 scoring function.

ML Python

selector = SelectKBest(score_func=[1], k=5)

Drag options to blanks, or click blank then click option'

Amutual_info_classif

Bf_classif

Cchi2

Dr2_score

Attempts:

3 left

💡 Hint

Common Mistakes

Using f_classif which is for ANOVA F-value, not chi-squared.

Using r2_score which is a regression metric, not a scoring function for feature selection.

The chi2 function is a common scoring function used with SelectKBest for classification tasks with non-negative features.

3fill in blank

hard

Fix the error in the code to correctly fit the feature selector to the data X and labels y.

ML Python

selector = SelectKBest(score_func=chi2, k=3)
selector.[1](X, y)

Drag options to blanks, or click blank then click option'

Afit

Btransform

Cpredict

Dfit_transform

Attempts:

3 left

💡 Hint

Common Mistakes

Using transform before fitting causes an error.

Using predict is not valid for feature selectors.

The fit method trains the selector to find the best features based on X and y.

4fill in blank

hard

Fill both blanks to create a dictionary of feature scores and select features with scores greater than 10.

ML Python

scores = {feature: score for feature, score in zip(feature_names, selector.scores_)}
selected_features = [feature for feature, score in scores.items() if score [1] [2]]

Drag options to blanks, or click blank then click option'

A>

B10

C<

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' selects low scoring features.

Using 5 instead of 10 changes the threshold.

This code selects features whose scores are greater than 10.

5fill in blank

hard

Fill all three blanks to create a new dataset X_new with selected features and print its shape.

ML Python

X_new = selector.[1](X)
print('Selected features shape:', X_new.[2])
num_features = X_new.[3]

Drag options to blanks, or click blank then click option'

Atransform

Bshape

Cshape[1]

Dfit

Attempts:

3 left

💡 Hint

Common Mistakes

Using fit instead of transform to get selected features.

Using shape[0] instead of shape[1] to get feature count.

The transform method applies feature selection to X. shape gives the dimensions of the array, and shape[1] gives the number of columns (features).